THE USE OF POLARIZED LIGHT IN VEGETABLE HISTOLOGY. 295 
secondary walls pass into the pore-canals ; while Theodor Hartig 
advances the opinion that it is the innermost layer of the cell-wall 
(the inner membrane of the newer, the tertiary 
membrane of the older authors) which is trans- I P 
formed into pore-canals, and that the closure is ; 
formed by two adjoining cells here brought to- 
gether (Fig. 3), which opinion I can confirm, after 
carrying out a series of researches. The fact, as 
represented in the annexed figure, is now also 
admitted by Strasburger, but another interpretation 
is given to it. According to this investigator, the 
inner layer is said to represent a later differentia- 
tion, which arises in consequence of contact with 
the cell-contents ; and the stronger light-refracting 
layer, which only apparently extends from the 
inner plane uninterruptedly into the pore-canal, 
is said to represent just such a differentiation of 
the secondary thickening in the parts adjoining 
the pore-canal, whilst the closed end of the pore 3 
is formed out of the primary walls. ; 
Whereas the collective layers of the cell-wall possess the form 
of an ellipsoid of elasticity, which, in the transverse and longitudi- 
nal sections, appears as a section of an ellipse, in which the smallest 
axis lies radially or perpendicularly to the stratification, the greater 
axis parallel to the stratification (the transverse section yielding the 
least, the longitudinal section the greatest axis), so observation with 
polarized light must afford the necessary and most trustworthy 
elucidation of the course of the stratification. If the course of the 
inner layer is in this way, as Hartig and I maintain, then the ar- 
rangement of the section falling through the ellipsoid of elasticity 
must change in the section-plane, just in the way in which it is 
represented in the schematic figure 4. If, on the contrary, the 
real structure is in accordance with the view of Strasburger, then 
such a change cannot take place, and the arrangement of the 
ellipsoidal section in the parts in question of the wall must be as 
represented in the schematic figure 5. 
If we now observe in the field of view (coloured red by means 
of a small selenite plate) a section sufficiently thin (so thin, indeed, 
that in the supposed arrangement of the experiment, the added 
colour will rise only to violet, the withdrawn colour fade only to 
orange) through the seed-albumen of //ytelephas macrocarpa, 
which object is one of the most suitable on account of the deep 
pore-canals, then the inner layer, like the other wall-layers, will 
appear in yellow light under an angle of+ 45° in the place where 
the inflection of the pore-canal lies ; therefore, where the axes of 
elasticity of the inner layer run parallel with the planes of polariza- 
